Advancements in cancer research reveal AI prediction tools and metastatic mechanisms

Researchers have announced two significant advancements in cancer treatment and diagnosis. An international team developed Spalmmune, an artificial intelligence model designed to predict patient responses to immune checkpoint inhibitors. In studies involving advanced melanoma and non-small cell lung cancer, the AI achieved a sensitivity of 92% and a specificity of 83%, outperforming conventional biomarkers by up to 30%. The model identifies the spatial proximity between B lymphocytes and follicular helper T cells within intratumoral tertiary lymphoid structures as a key predictor of therapeutic success.

In a separate study, researchers have identified a molecular mechanism that could help combat breast cancer metastasis. The investigation focused on the protein RAC1 and its role in tissue invasion. Scientists discovered that a chemical modification called SUMOylation acts as a molecular switch, allowing RAC1 to interact with the protein POTEE. This interaction facilitates the formation of invadopodia, structures that enable tumor cells to degrade surrounding tissue and spread to other organs. The goal of this research is to develop strategies to prevent this specific protein activation to block cancer progression.